metadata
library_name: transformers
language:
- en
license: apache-2.0
base_model: answerdotai/ModernBERT-base
tags:
- generated_from_trainer
metrics:
- precision
- recall
- f1
model-index:
- name: DisamBertSingleSense-omsti
results: []
DisamBertSingleSense-omsti
This model is a fine-tuned version of answerdotai/ModernBERT-base on the semcor dataset. It achieves the following results on the evaluation set:
- Loss: 3.1672
- Precision: 0.7233
- Recall: 0.7064
- F1: 0.7148
- Matthews: 0.7058
Model description
More information needed
Intended uses & limitations
More information needed
Training and evaluation data
More information needed
Training procedure
Training hyperparameters
The following hyperparameters were used during training:
- learning_rate: 0.0001
- train_batch_size: 64
- eval_batch_size: 64
- seed: 42
- optimizer: Use OptimizerNames.ADAMW_TORCH_FUSED with betas=(0.9,0.999) and epsilon=1e-08 and optimizer_args=No additional optimizer arguments
- lr_scheduler_type: inverse_sqrt
- lr_scheduler_warmup_steps: 1000
- num_epochs: 5
Training results
| Training Loss | Epoch | Step | Validation Loss | Precision | Recall | F1 | Matthews |
|---|---|---|---|---|---|---|---|
| No log | 0 | 0 | 88.3567 | 0.4162 | 0.3628 | 0.3877 | 0.3621 |
| 2.0504 | 1.0 | 17769 | 3.1969 | 0.7332 | 0.7147 | 0.7238 | 0.7142 |
| 1.8615 | 2.0 | 35538 | 3.3690 | 0.7334 | 0.7130 | 0.7230 | 0.7124 |
| 1.8386 | 3.0 | 53307 | 3.2408 | 0.7246 | 0.7064 | 0.7154 | 0.7058 |
| 1.8802 | 4.0 | 71076 | 3.2913 | 0.7337 | 0.7121 | 0.7227 | 0.7115 |
| 1.7798 | 5.0 | 88845 | 3.1672 | 0.7233 | 0.7064 | 0.7148 | 0.7058 |
Framework versions
- Transformers 5.2.0
- Pytorch 2.10.0+cu128
- Datasets 4.5.0
- Tokenizers 0.22.2